Consociationalism and Racial Cleavages: Redefining the Boundaries of Consociationalism

By Jitske Mijna Grift

Submitted to

Central European University

Department of Political Science

In partial fulfillment of the requirements for the degree of Masters of Arts/Sciences

Supervisor: Professor Matthijs Bogaards

CEU eTD Collection

Budapest, Hungary

(2019)

Abstract

“A portion of mankind may be said to constitute a Nationality if they are united among

themselves by common sympathies which do not exist between them and any others—which

make them co-operate with each other more willingly than with other people, desire to be

under the same government, and desire that it should be government by themselves or a

portion of themselves exclusively.” (Mill, 1873, 308). With this, John Stuart Mill opened

chapter sixteen of his book Considerations on Representative Democracy, and established his

case for why democracies need to have a uniting factor. But what about countries that do not

have this uniting factor? Countries that have divides. Almost 200 years after John Stuart Mill

published his book, Arend Lijphart wrote about just that, democracies in divided societies.

Lijphart coined the theory consociational democracies, which are democracies that have

divides based on factors such as language, religion, ethnicity, race, or culture, but they still

function as democracies (Lijphart, 1969). However, Lijphart’s theory has not been without

controversy, as some criticize the idea that consociationalism can work for countries that have

a racial divide (Barry, 1975). The question about whether democracy can work in racially

divided societies is now more relevant than ever, with globalization and international

migration, more and more societies are becoming racially diverse. This paper will look deeper

into the question of how consociational societies handle racial divides, and by doing that

hopes to show that an consociationalism is incompatible with a racial divide.

CEU eTD Collection

Acknowledgments

I would like to thank several people for their help in creating this thesis.

To Mahnoor, thank you for putting up with me during this writing process.

To Anashua, you didn’t help but I knew you would be upset if I did not include you.

To Kinga, in vino sapientia

To Viswesh, thank you for your conversations, they have helped me greatly

To my little sister Mijke, who always picked up the phone even when it was late at

night, I love you and I miss you. I’ll be home soon.

A special acknowledgment goes to my parents, Gert Grift and Thera Wolf, who have

fought hard their entire lives to give me the opportunities that did not come easy to them

growing up. If it wasn’t for their education, their determination, and their incredible working

ethic, I would not be where I am right now.

This thesis is dedicated to my grandparents, who were forced to stop going to school

at a young age, who had to work hard their entire lives, who never got to experience their

childhood as I did. I did it for all of you.

“En aan de leraar die mij altijd placht te dreigen

Jongen jij komt nog op het verkeerde pad CEU eTD Collection Kan tevreden zijn en hoeft niet meer te krijgen

Dat wil zeggen, hij heeft toch gelijk gehad”

- Testament, Boudewijn de Groot

Table of Contents Abstract ...... i

Acknowledgments ...... ii

List of figures ...... iii

Introduction ...... 1

Chapter 1 ...... 6

1.1 Research questions ...... 6

1.2 Definitions ...... 8

Chapter 2 ...... 16

2.1 Data ...... 16

2.2 Consociational democracy or not? ...... 27

Chapter 3 ...... 31

3.1 The end of consociationalism?...... 31

3.4 Conclusion ...... 44

Sources ...... 50

CEU eTD Collection

List of figures

Table 1……………………………………………………………………………..17

Table 2……………………………………………………………………………..27

CEU eTD Collection

iii

Introduction

In his paper Consociational Democracy, published in 1969, Arend Lijphart called

attention to a type of democracy that he found to have gone unnoticed. Lijphart mentions the

classification of political systems by Gabriel A. Almond, who classifies political systems in

three different categories (Lijphart, 1969, 207). First, he mentions the political system in

countries such as Britain and the US, which he classified as the Anglo-American political

system (Lijphart, 1969, 207) Second, Almond mentions the political system in France

Germany and Italy, which he classified as the Continental European political system

(Lijphart, 1969, 207). The third category was not specified by Almond, but includes the Low

Countries of Belgium, Luxembourg, and the Netherlands, and the Scandinavian countries

(Lijphart, 1969, 207). Almond does not go deeply into the political system of the third

category, instead saying that they are a combination between both the Continental and Anglo-

American political system (Lijphart, 1969, 207). While the qualifications have specific

geographical names, Almond’s classifications were not tied to any geographic location, as

Lijphart mentions (Lijphart, 1969, 207-208). Lijphart focuses his paper on the third

classification that Almond made, the Low Countries and Scandinavia. While Almond claims

that these countries are hybrids of the two other categories, Lijphart claims that these

countries are actually their own separate political system, which he names as consociational

democracies (Lijphart, 1969, 207). Lijphart describes consociational democracies as the

following; “Consociational democracy means government by elite cartel designed to turn a

CEU eTD Collection democracy with a fragmented political culture into a stable democracy.” (Lijphart, 1969, 216).

This is a very concise way of describing consociationalism, so perhaps a broader description

might be better. Consociationalism is a political system for divided societies, this divide

might be caused by race, religion, language, ethnicity, or race, and the divide is also seen in

the political system, where political parties represent their own segment of society (Bogaards,

2017, 1). Consociationalism focusses on the political leaders of the different segments of

society, who, realizing that they will have to cooperate to run a stable country, decide to

accommodate each other (Bogaards, 2017,1). Consociational democracy is built on four

principles, which were described by Bogaards in his entry into the Wiley Blackwell

Encyclopedia of Social Theory, which are the following; “a grand coalition government with

the leaders of all main parties/communities; proportionality in political representation and the

distribution of resources; segmental autonomy; and a mutual veto” (Bogaards, 2017, 1).

However, these principles are not absolutely set, they can be adapted to the country that

decides to use consociationalism to deal with the divisions in the society (Bogaards, 2017, 1).

Lijphart’s model was used around the world to bridge divides in society and end conflict

(Bogaards, 2017, 1).

While consociationalism was embraced by parts of the political science community as

a way to handle conflict in divided societies, there was also criticism against the theory. Part

of this criticism was against the use of consociationalism in societies divided by racial

conflict. In his paper, Political Accommodation and Consociational Democracy, Brian Barry

writes about his criticism against consociationalism. While he first discusses how he does not

believe all countries that are claimed to be consociational are consociational, he later goes

further into how consociationalism is not compatible with racial conflict, however Barry

refers to this as ethnic conflict. Barry mentions four reasons why consociationalism could not

be compatible with ethnic divisions. The first is the fact that acts of gross inhumanity are CEU eTD Collection mostly aimed at groups that are ethnically different from the perpetrators, especially when the

victims also have physical or cultural differences (Barry, 1975, 502). The second is that

religion and class different are about organizations, specifically belonging to a certain

organization, whereas ethnic differences are about solidarity groups, while they might have an 2

organization, it is not necessary to have an organization to start riots just the ability to

recognize who belongs to which groups (Barry, 1975, 502). Third, is the fact that ethnic

groups do not have a set way of interpreting the world, religion can follow a certain set of

rules which can be used by political leaders to explain why a policy is necessary, but that is

not true for an ethnic group (Barry, 1975, 502). The fourth is perhaps the most important one,

a religious or class conflict is about how the country is run, what values are more important

for instance, however an ethnic conflict may not be about how the country is run, but if it

should be a country at all (Barry, 1975, 503). This is important because it is something that

cannot easily be solved by cooperation nor accommodation, as it is about the fundamental

existence of the country itself.

Another argument regarding racially divided societies comes from Donald Horowitz,

whose argument mentions ethnicity but is also applicable to a racially divided society. In his

article Three Dimensions of Ethnic Politics, Horowitz discusses the difference between

horizontal and vertical ethnic differentiation (Horowitz, 1971, 232). Systems that have

horizontal ethnic differentiation have parallel ethnic structures, while the question of group

superiority can still exist, generally ethnic groups in horizontally ethnic differentiated systems

do not have a definitive hierarchy between ethnic groups (Horowitz, 1971, 232). Systems that

have vertical ethnic differentiation are almost the opposite of this, they have a hierarchy

between ethnic groups, and this hierarchy influences things such as politics and social

mobility (Horowitz, 1971, 232). The best example of a system of vertical ethnic

differentiation is that of Apartheid South Africa, where the white minority was considered CEU eTD Collection superior to the non-white majority, while this is an extreme example there have been other

systems of vertical ethnic differentiation as well, such as segregation in the US, which is an

example Horowitz uses as well (Horowitz, 1971, 233). While Horowitz’s piece is not

necessarily critical of consociationalism, he never mentions consociationalism in his article, it 3

is a good way of looking at different societies that have race as their main divide, and

therefore it is useful for this research as well.

Rabushka and Shepsle also provided criticism of consociationalism in ethnically

divided societies. In their book, Politics in Plural Societies: A Theory of Democratic

Instability, the authors describe several cases where ethnically diverse societies experienced

conflict. Rabushka and Shepsle question how consociationalism can solve these conflicts,

especially because many of their cases experienced civilized power-sharing at one point, and

violent conflict at another (Rabushka and Shepsle, 1972, 207-208). They offer several

solutions that could be used to deal with ethnically divided societies, although they argue that

the feasibility of these solutions is not guaranteed (Rabushka and Shepsle, 1972, 213). The

first solution is the “Denial of independent, decision-making authority” (Rabushka and

Shepsle, 1972, 213). With this they show that leaders from different race groups can work

together in times of colonial conflict, but not when the colonial powers are gone, by not

allowing these countries to be independent, the ethnic conflict will not come up (Rabushka

and Shepsle, 1972, 213-214). The second point is “Restrictions on independent, decision-

making authority” (Rabushka and Shepsle, 1972, 215). This means decentralizing the

government and putting most decisions on the local level not the federal level (Rabushka and

Shepsle, 1972, 215). The third point is “Restrictions on free political competition” (Rabushka

and Shepsle, 1972, 215). With this, the authors argue, the elites would practice a level of

secrecy when it comes to policy making, and disregarding the pressure of the mass electorate

(Rabushka and Shepsle, 1972, 215-216). The fourth solution is “Restrictions on the scope of CEU eTD Collection government” (Rabushka and Shepsle, 1972, 216). By taking away the government’s ability to

distribute public goods, they argue, the reason for etnnic conflict will also be diminished, as

distributing public goods can lead to the government giving more than one race over the other

(Rabushka and Shepsle, 1972, 216). The fifth solution is “Creation of homogenous societies” 4

(Rabushka and Shepsle, 1972, 216). This would mean a form of ethno-nationalism, allowing

countries to break up in accordance with ethnicity (Rabushka and Shepsle, 1972, 216-217).

The sixth solution is “Creation of permanent external enemies” (Rabushka and Shepsle, 1972,

217). This would mean uniting the people against a common enemy, forcing them to work

together through that (Rabushka and Shepsle, 1972, 217). However, while the authors offer

many solutions, they also critique the solutions they give, seeing most of them as not

completely viable, and some as going directly against the democratic process. In the end, they

conclude, that there is no way a society with intense differences can be manageable, painting

a bleak picture for the countries that have racial difference (Rabushka and Shepsle, 1972,

217).

The debate regarding consociationalism and its success for multiracial societies is still

open, and in his book Power-Sharing in South Africa, Arend Lijphart responds to his critics as

well. While this research might not be as influential as the book by Rabushka and Shepsle or

the response from Arend Lijphart, it would like to add to that discussion. By looking deeper

into racially divided societies, looking at how they function, and if they are even still

consociational.

CEU eTD Collection

Chapter 1

1.1 Research questions

As was presented above, the criticism against consociationalism is not rare, and

specifically the criticism against using consociationalism for racial conflict. But with the

world becoming more globalized, there are more and more societies that have racial

cleavages. The Netherlands, the country that Lijphart used as his first model on

consociationalism, has also started experiencing racial differences (Bogaards, 2017, 1). The

country took in guest workers in the 1960’s and 1970’s fro countries such as Morocco,

Turkey, Spain and former Yugoslavia (De Valk, Esveldt, Henkens, Liefbroer, 2001, 50). At

the same time the Netherlands also lost its colony of Surinam, which led to immigration from

Surinam to the Netherlands as well (Biervliet, Bovenkerk, Köbben, 1975, 337). This led to the

Netherlands currently having 1.2 million people who are decedents of immigrants from

Suriname, Turkey, Morocco, or the Dutch Antilles (CBS, 2017). With the arrival of new

immigrants, the Netherlands also experienced a growth in its Muslim population. In 2017,

about five percent of all citizens in the Netherlands identify as Muslim (NOS, 2017). This

added a new religion to the power sharing structure in the Netherlands, which was based on

different denominations of Catholics, and Protestants, and a general power which mostly

consisted of Liberal and Socialist (Lijphart, 1990 ,96). One would assume that, because the

CEU eTD Collection societal cleavages were based on religion, the Muslim power would easily integrate into the

power sharing structure, but this was not the case. There have been many lawsuits about

Muslim education in the Netherlands, while there are parties of different denominations in the

Dutch parliament, there is no Muslim party (Driessen, Merry, 2006, 204) (Tweede Kamer der

Staten-Generaal, 2019). The question here is, is it really about the Muslim religion? Or would

things have been different if the growth of the Muslim population came from the conversion

of white Dutch people? Of course, there has been a lot of criticism regarding racial conflict

and consociationalism already, some of which was mentioned earlier. But this paper would

like to look beyond just the theoretical approach of racial conflict and consociationalism, and

look at different consociational or formally consociational countries. By looking at

consociational and former consociational countries this paper would like to show that

countries that have mainly an racial divide, instead of a religious or class divide, are more

likely to have conflict, and could even be more likely to fail at implementing a consociational

democracy. By looking at not just success cases of consociationalism, but also look at cases

that failed, we can see what exactly made consociationalism fail in different countries, and

how this can be prevented for other countries, or how we can see if a country is about to fail

as a consocational democracy. Of course, this paper does not advocate for the idea that people

of different races cannot live together in one society, but instead would like to show how, if a

society does have different races, one can spot when consociationalism is failing. In order to

do this, this paper will look at two different research questions. The first is, are societies that

are racially divided and claimed to be consocational still consociational? And if not, what is

the reason that these societies are no longer consociational? This paper hypothesizes that most

racially divided societies are no longer consocational, and the main reason for that is the racial

divide in the society.

CEU eTD Collection

1.2 Definitions

In order for the question of consociationalism and racial cleavages in a society to be

studied, first we must establish some definitions for different concepts. The first concept is

that of race, and specifically the difference between race and ethnicity.

The question on why a definition is necessary is not something that has not been asked

before. In her paper on ethnicity, Kanchan Chandra asks the question of what ethnicity is, and

why it matters (Chandra, 2006). First, she begins her paper by asking why a definition for

ethnicity is even important, and quickly answers it as well (Chandra, 2006, 398). Chandra

argues that a definition for ethnicity is important because it allows us to build theories about

ethnic identity and concepts based on ethnic identity (Chandra, 2006, 398). This is also the

case for this research, without giving a definition for race, we cannot look at what countries

have different racial groups in the first place. Chandra argues that ethnicity is part of the

identity categories that are descent based, meaning that what ethnicity someone belongs to is

in large part decided by their ancestors (Chandra, 2006, 399). This is clear by the two

properties she finds to be intrinsic to ethnicity, constrained change and visibility (Chandra,

2006, 399). With constrained change, Chandra means that, while someone’s ethnicity can

change in the short term, it is constrained by an underlying set of attributes (Chandra, 2006,

399). When it comes to visibility, Chandra means that some information of someone’s

ethnicity can be found through observation, meaning looking at someone’s hair or facial

figure (Chandra, 2006, 399). The question that we now must pose is if these are factors that

not just influence ethnicity but also race? After all, there are cases of people who share similar CEU eTD Collection features and descent, but are still seen as belonging to different ethnic groups. Some scholars

argue that common ancestry or a myth of common ancestry, a common region of origin or a

myth of a common region of origin, what ethnic group one’s parents were assigned to, a

common culture, a common history, or a combination of characteristics, are also important

when it comes to the definition of ethnicity (Chandra, 2006, 403-414). In her paper, Chandra

provides examples of cases that do not fit those definitions of ethnicity, and thereby shows

that none of those definitions are broad enough to include all types of ethnic groups (Chandra,

2006, 403-414). Chandra argues that the definition of ethnicity itself is broad, simply because

ethnicity itself is broad, but that does not mean that you cannot define ethnicity (Chandra,

2006, 421-422). Instead, she argues, the definitions that are giving to ethnicity are secondary,

they can be added to ethnicity when necessary, but are not the main definition of ethnicity

(Chandra, 2006, 413). The main point that Chandra tries to make with her paper is that

ethnicity itself is not important, but often used for things such as ethnic violence or ethnic

conflict, however ethnicity itself does not make a difference (Chandra, 2006). So if ethnicity

is not important, as Chandra states, is it not the same with race? Why should we use race at

all in this paper? Just like ethnicity, race is in itself not important. However, just like Chandra

argues about ethnicity, race is important when it comes to other things, such as racial conflict

and discrimination. Race is not the dependent variable in this paper, we are not testing race

when it comes to democracy. Instead we are looking at if consociationalism is possible when

it comes to countries that have a racial divide, something that has often been argued against.

We will now go further into the criticism against using consociationalism in racially divided

societies, but will also keep in mind Chandra’s two main properties of ethnicity that are

similar when it comes to race, constrained change and visibility, to use for our definition.

In his paper on consociationalism, Barry explains why he believes that CEU eTD Collection consociationalism does not work for societies where the main dividing factor is ethnicity. For

this, Barry gives four different reasons, the fourth one was already mentioned earlier in this

paper, but the first three reasons are more important when it comes to the establishing of a

definition for ethnicity.). Barry refers to ethnicity as belong to a ‘people’ or a ‘race’, but 9

perhaps what is most important in his argument is the fact that an ethnic group can easily be

distinguished from others (Barry, 1975, 503).

The first reason Barry gives is that horrible violence, or as Barry calls it “acts of gross

inhumanity”, are more easily carried out or supported if the victims are of a different ethnic

class (Barry, 1975, 502). The second reason Barry gives is that ethnicity, unlike religion or

class, does not need an organizational structure to carry out violent acts such as riots (Barry,

1975, 502). As long as they have a way of distinguishing between those who belong to the

same ethnicity and those who do not, an organizational structure is not needed (Barry, 1975,

502). The third reason Barry gives is that there is no theoretical argument needed for someone

to claim that they know what is best for their ethnic group (Barry, 1975, 502). A religious

leader might need to refer to specific religious texts, but an ethnic leader does not, all it needs

is the support of the group (Barry, 1975, 502). The fact that ethnic groups are easily

distinguishable is the main argument that Barry has against using consociationalism for

ethnically divided societies.

The link that Barry makes between ethnicity and distinguishable features is one that is

also made by Rabushka and Shepsle in their book on consociationalism. Rabushka and

Shepsle use four different indicators for ethnicity: race, religion, language, and tribe and

custom (Rabushka and Shepsle, 1972, 8-10). They argue that these are all ethnic divisions,

and that the ethnic divisions coincide with political divisions (Rabushka and Shepsle, 1972,

10). However, the authors consider some indicators of ethnicity more important than others.

When it comes to the religion indicator and the language indicator, the authors argue that both CEU eTD Collection are part of a larger ethnic division and are not the only indicators of different ethnicities

(Rabushka and Shepsle, 1972, 9-10). As an example for this they discuss the case of Belgium

when it comes to a language division; “For example, Flemings and Walloons in Belgium each

insist they are the product of a long history of different cultural experiences of which 10

language is only a surface characteristic.” (Rabushka and Shepsle, 1972, 9). They argue that it

is the same way for religion, religion can be a divide in society, but only when related to a

larger divide, and a common religion also does not mean there is no divide in a society

(Rabushka and Shepsle, 1972, 9-10).

But the criticism against consociationalism has not gone unanswered, Lijphart has

reacted specifically to Barry’s criticism in his book “Power-Sharing in South Africa”

(Lijphart, 1985). In it Lijphart addresses the different points that Barry made when it came to

consociationalism within ethnically divided societies. When it comes to the organizational

structure of class or religion, as opposed to the lack of an organizational structure in within an

ethnic group, Lijphart argues that ethnic groups organize themselves just as well as any other

group within a plural society (Lijphart, 1985, 96). Lijphart finds that Barry mainly questions

the elite control of the different groups, and not on the reaching of agreements across different

groups (Lijphart, 1985, 96). When it comes to reaching agreements across different groups,

Lijphart argues, this is done easier between ethnic groups, because they do not have religious

and ideological differences unlike different religious or class groups (Lijphart, 1985, 96).

Lijphart argues that, because religious and class differences are logically based, on the

interpretation of religious text for example, they are less easy to compromise, whereas ethnic

differences are emotionally based, which makes them easier to find a common ground

(Lijphart, 1985, 96). Barry disputed this in his paper, specifically because he argued that when

it comes to religious or class differences, this is mainly about how to run a country, whereas

ethnic differences are about if the country should exist in the first place, or at least who should CEU eTD Collection be in it (Barry, 1975, 502). Lijphart argues against this as well, claiming that ethnic groups

only dispute the existence of the country when they are geographically separated as well, not

just ethnically separated (Lijphart, 1985, 96). A final argument that Lijphart makes is that

ethnicity is difficult to distinguish from religion, for which he gives the examples of Northern 11

Ireland and Lebanon, which have often been seen as not just religious but also ethnic divides

(Lijphart, 1985, 97). This final point is especially something to avoid in this paper, as it is true

that religion and ethnicity can often be linked, but in Northern Ireland and Lebanon, the

differences are not based on phenotypical features, mainly the difference between a catholic

or a protestant in Northern Ireland is not something that can be recognized by looking at hair

types or facial features, something which can be defined as being part of a different racial

group.

Another reaction to Barry’s criticism is that of Brendan O’Leary, in his chapter of the

book “From Power Sharing to Democracy: Post-Conflict Institutions in Ethnically Divided

Societies.” O’Leary dissects Barry’s criticism into two different statements, first he mentions

Barry’s argument that ethnic division are mostly about whether a state should be a state and

not about how the country is run, to which O’Leary agrees that conflict is less mendable if

separatism is involved (Noel, 2005, 26). O’Leary also agrees that class conflict is less violent

than ethnic conflict, as Barry stated that ethnic conflict is much more likely to lead to violence

(Noel, 2005, 26). However, O’Leary does not agree with Barry fully, he argues that

separatism is not always involved in ethnic conflict, and if the different ethnicities agree on

the existence of the state and its integrity, then consociationalism is possible, for which he

uses the example of Belgium and Switzerland (Noel, 2005, 26). O’Leary argues that ethnic

communities might be easier to find a compromise than religious communities, because they

are not as divided by ideology (Noel, 2005, 27). O’Leary also argues that it is rare for states to

not have multiple cross-cutting cleavages, such as intersections between race and class, and CEU eTD Collection that pluralist would argue that these cross-cutting cleavages actually dampen one another

(Noel, 2005, 27). Of course, O’Leary argues, there are cases where cross-cutting cleavages

actually strengthen one another, for example a party alliance based on race that is also about

class, in those cases consociational institutions can help (Noel, 2005, 27). However, O’Leary 12

also mentions that when one group holds a disproportionate amount of economic power, than

conflict is more likely and that conflict is more likely to be violent, which is why Lijphart

argued that having near socio-economic equality is a condition for consociationalism to work

(Noel, 2005,27). Does this mean that consociationalism can only work in societies that have a

moderate divide? O’Leary argues against that, he finds that, while it is true that

consociationalism is easier in societies that have a moderate divide, it is not impossible to use

consociationalism in societies that have deeper divides, for which he uses the example of

Northern Ireland, South Africa, and Lebanon (Noel, 2005, 28).

The response that Lijphart and O’Leary gave to Barry makes a lot of good points, but

they both seem to be missing a crucial part of Barry’s criticism. Mainly, that Barry bases his

criticism mostly on the ethnic groups that are easily distinguishable. Lijphart and O’Leary

both pointed out countries that have ethnic differences, but not easily distinguishable ones,

such as Switzerland, Belgium, Lebanon, and Northern Ireland. The distinguishable features

are very important when it comes to Barry’s criticism, and it is also an important part of

Rabushka and Shepsle’s criticism of consociationalism, however they refer to the

distinguishable features as phenotypical features and mention it as being a part of race

(Rabushka and Shepsle, 1972, 8). This paper will look specifically at countries that have

groups that have different distinguishable features, such as hair, skin colour, and facial form,

because these are the types of countries that are incompatible with consociationalism

according to critics, and just as Rabushka and Shepsle, this paper will refer to those

differences as race and not as ethnicity. This is to avoid the same mistake that Lijphart and CEU eTD Collection O’Leary made, mainly that people can belong to a different ethnicity, as in Lebanon or

Northern Ireland, and still have the same phenotypical features. It is therefore better to refer to

these differences as racial, so that this research will not include countries that have different

ethnic groups that do not have different phenotypical features, as this will best fit the criticism

against consociationalism that was provided by Barry and Rabushka and Shepsle.

However, it is important to mention here that when talking about these different

features, or different ethnic groups, this paper does not advocate for the idea of primordialism,

which is the idea that there is a relationship between people who belong to the same ethnic

group solely based on their shared ethnicity (Grosby, 2010). In fact, like many critics of

consociationalism, this paper would argue that consociationalism itself does this, it makes

people see themselves as having a specific relationship with their racial group which, unlike

religion or class, has no common ideology to unite people. But because of the importance of

race, when it comes to the criticism of consociationalism, this paper will use the following

definition of race; Belonging to a ‘people’ that have distinct phenotypical features, religion or

language can be part of race but are not the main distinguishing factor.

The second concept that is necessary to define, is the concept of ethnic conflict.

Conflict is difficult to define, as it can involve many different things. Conflict can mean

violence, as in an armed conflict, or it can be a conflict on an individual level, between two

individuals. However, the use of ethnic conflict in this research will be one on a national

level, between groups in society. Of course, there is also differences between conflict, and

ethnic conflict is just one type of conflict out of many. However, because this research looks

specifically at societies that have conflict because of an ethnic divide, we will only look at

ethnic conflict. When it comes to the use of violence in conflict, this paper will look at both

violent and non-violent conflict. While violent conflict shows and obvious failing of the CEU eTD Collection consociational democracy, non-violent forms of conflict also show that consociationalism has

not worked, as consociationalism is about managing any form of conflict between groups. For

the definition of ethnic conflict, this research will look at the definition provided by Karl

Cordell and Stefan Wolff. Cordell and Wolff describe conflict as “a situation in which two or 14

more actors pursue incompatible, yet from their individual perspective entirely just, goals.”

(Cordell and Wolff, 2011, 4). However, ethnic conflict goes a little further than that, as

Cordell and Wolff put it, ethnic conflict is “that in which the goals of at least one conflict

party are defined in (exclusively) ethnic terms, and in which the primary fault line of

confrontation is one of ethnic distinctions.” (Cordell and Wolff, 2011, 4). This means that the

conflict itself must be motivated by ethnicity by at least one party, and that the main part of

the conflict itself is related to ethnicity. By taking from the definitions of conflict and ethnic

conflict provided by Cordell and Wolff, this research shall use the following definition of

ethnic conflict; the incompatibility of goals between two or more actors, where at least one of

the actors’ goals are based in ethnicity and where the main part of the conflict is motivated by

ethnicity.

CEU eTD Collection

Chapter 2

2.1 Data

The countries that will be used in this study are selected from a list first created by

Paul Dixon. Dixon made a list of countries that were claimed to be consociational and put

them in a table, depending on when and by whom they were claimed to be consociational

success stories (Jakala, Kuzu, Qvortrup, 2018, 68-69). The list comes from the book Thinking

about Democracy: Power Sharing and Majority Rule in Theory and Practice, by Arend

Lijphart (Jakala, Kuzu, Qvortrup, 2018, 68-69). The list itself contains all countries that are

considered consociational by Lijphart, as of 2007 which is when the book was published

(Jakala, Kuzu, Qvortrup, 2018, 68-69). First, let us look at the list of countries that are

included. These are the following; Afghanistan, Antilles (NL), Austria, Belgium, Bosnia,

Burundi, Canada, Colombia, Cyprus, Czechoslovakia, Fiji, India, Israel, Kosovo, Lebanon,

Luxembourg, Malaysia, Macedonia, Netherlands, Nigeria, Northern Ireland, South Africa,

Suriname, Switzerland, Uruguay (Jakala, Kuzu, Qvortrup, 2018, 68-69).

Because this research is based on consociational democracies that have racial divides,

we will first look at what type of divide each country has, racial or non-racial. We will do this

by going over each country and using the definition of race provided above to establish

whether the divide is racial. The following table shows the findings from this research and

following will be the explanation for each country as to why it has does not have a racial

divide. CEU eTD Collection

Table 1

Afghanistan is an incredibly diverse country when it comes to tribal backgrounds with

forty two percent of the country being Pashtun, and the rest of the country being compromised

by Tajiks, Uzbeks and Hazaras, and other small tribes. (Adeney, 2008, 538). However, the

main divide in Afghan society is not a tribal one, but a religious divide, mainly the divide CEU eTD Collection between the extremely conservative Muslims, the less conservative Muslims, and the different

types of sects within Islam (Mishali-Ram, 2011, 264-268). While the religious divide and the

tribal divide are largely tied together, the main divide is an inter-religious divide within the

Islamic religion. Therefore, we will not consider Afghanistan to have a racial divide. 17

The Dutch Antilles were an autonomous territory within the Dutch kingdom (van

Aller, 1994, 575). A former colony of the Netherlands, the territory became semi-independent

in 1948 (van Aller, 1994, 573-574). The divide in the Dutch Antilles was largely based on the

competition between two of the islands, Aruba and Curacao (van Aller, 1994, 574). Aruba

wanted to be independent of Curacao and instead have its own independent relationship with

the Netherlands (van Aller, 1994, 574). This came from the fear of Aruba that Curacao would

become too powerful within the island group, Curacao’s population was bigger than that of

Aruba, however Aruba was more densely populated (van Aller, 1994, 577). Nevertheless, the

divide within the Dutch Antilles was not a racial divide, but one of nationality, a competition

between two different islands. Therefore, we will not consider the Dutch Antilles to have a

racial divide.

Austrian society has several cleavages which play out into political cleavages. First

there is the economic cleavages in Austrian society, between the upper class and the working

class (Hafez and Heinisch, 2018, 652). Second there is the religious cleavage, which is

between Catholics and seculars (Hafez and Heinisch, 2018, 652). These cleavages play out in

the political arena with party association, the working-class usually votes for the Socialists,

while the upper class votes for the People’s Party (Bingham Powell, 1976, 3). While this

means that Austrian society has a religious, political, and class divide, it does not have an

racial divide.

Just as Austria, Belgium has several cleavages, the most noticeable one being CEU eTD Collection language. Belgium is divided up in two main language groups, with the Flemish speaking part

in the North, and the French speaking part in the South (Deschouwer, 2012, 8). But language

is not the only cleavage that exists in Belgian society, there is also a divide between the

Socialists and the Catholics, with the Catholics being more represented in the Northern 18

Flemish speaking part, and the Socialists being more represented in the Southern French

speaking part that housed more industrial areas (Deschouwer, 2012, 8). However, while the

French and Flemish communities might be separate in a lot of ways, they do not belong to

different racial groups. The two groups do not have different phenotypical features, and

therefore they do not fit our description of different racial groups. Therefore, Belgium will not

be considered as having a racial divide.

The divide in Bosnian society is mainly between the Bosnian Muslims, Bosnian Serbs,

and Croats (Dahlman and Tuathail, 2005, 575). This divide resulted in a conflict during the

Bosnian war in the 1990’s, with the Bosnian Muslims being caught between the Bosnian

Serbs who wanted Bosnian Serb regions to be a part of Serbia, and the Croats who wanted

their regions to be part of Croatia (Dahlman and Tuathail, 2005, 575-577). However, while

the conflict in Bosnia was horrible, it was not a racial divide. The Bosnian Muslims and

Bosnian Serbs do not have different phenotypical features, and therefore do not fit our

description of different racial groups. Therefore, Bosnia will not be considered as having a

racial divide.

Burundi’s main divide is a tribal one. The country has two main tribes, the Hutus and

the Tutsis (Lemarchand, 2006, 7). The ethnic divide has led to violent conflict in the past as

well, with over 100,000 Hutus being killed in 1972 by the Tutsi controlled army (Uvin, 1999,

258). More violent conflict took place over the years, but went down after an attempt at

democratization in 1990, however after a coup that killed the democratically elected leader in

1993, the violence returned, with both sides killing each other (Uvin, 1999, 261-262). The CEU eTD Collection democratization process also brought with it two political parties that each represented one

ethnicity, although not officially, the Frodebu represented the Hutus, and the Uprona the

Tutsis (Lemarchand, 2006, 8). Nevertheless, what is clear is that the divide in Burundi

society is an ethnic one, not a racial one, as the Hutu’s and Tutsis do not have different

phenotypical features, and therefore Burundi will not be considered to have a racial divide.

Canada, like Belgium, has a language divide in its society. Canadian territory is

divided up between a French-speaking part and an English-speaking part (Cannon, 1982, 52).

While the divide in Canadian society can play out in religious, regional, and cultural, divides

as well, it is mainly a language divide (Cannon, 1982, 52). Therefore, Canada will not be

considered as having a racial divide, but instead a language divide.

Colombia’s case is one that can be summarized quickly. The divide in Colombia is not

racial, not religious, and not even class based, but instead a political divide between the

Liberals and the Conservatives (Dix, 1980, 304). The divide led to what was called a ‘quasi-

civil war’ in the 1940’s, and cost the lives of over 100,000 Colombians (Dix, 1980, 304).

Nevertheless, while the divide ran deep, it is not a racial divide, but still just a political divide.

The main divide in Cypriot society is between the Greek Cypriots and the Turkish

Cypriots, with the Greek Cypriots making up about eighty percent of the population (Jakala,

Kuzu, Qvortrup, 2018, 157-158). There are different accounts on how the conflict between

the two groups started, but what is known is that the Greek Cypriots wanted independence

from the British after the second world war, and become a part of Greece, the Turkish

Cypriots saw this as a threat as they were in the minority and sought for a dividing up of the

island into two separate territories (Jakala, Kuzu, Qvortrup, 2018, 158-159). The question of

whether the divide in Cyprus is a racial one is difficult, the question is if the phenotypical

difference between the Turkish Cypriots and the Greek Cypriots is enough to be considered CEU eTD Collection racial according to the definition provided. While this is an argument that can be made, this

research does not consider the Turkish Cypriots and the Greek Cypriots to have clear enough

different phenotypical features for the divide in Cypriot society to be considered racial.

Czechoslovakia is another country where the divide in society runs along national

lines. The Czechoslovak divide existed between the Slovaks, two different nationalities that

lived united under the Czechoslovakian government, but who also had large amounts of

autonomy (Macek-Macková, 2011, 620). The country ended up being divided in 1992, after

political leaders from both countries seemed to differ on what they saw as the future of the

country, and how to handle things such as the economy (Macek-Macková, 2011, 620). The

divide was not a violent one, but done through a referendum and political negotiations

(Macek-Macková, 2011, 620). Nevertheless, while the country technically no longer exists,

the divide in Czechoslovakia was one between two different nationalities, not an racial one.

Fiji’s divide is clearly a racial one. The Fijian society consists of two main racial

groups, indigenous Fijians, and Indian Fijians, and one smaller group that consists of other

racial groups such as Europeans, Part-Europeans, and Chinese (Milne, 1975, 414). The Indian

Fijians are descendants of indentured laborers who were send to Fiji when it was still under

British rule (Iyer, 2007, 132). The two communities are divided in multiple ways, which

includes different phenotypical features, but also language, customs, religion, and culture

(Iyer, 2007, 132). The divide is also played out in the economic sphere, where the Indian

Fijians might have started as indentured laborers, they ended up economically more powerful

than the indigenous Fijians, which has led to a fear within the indigenous Fijian community of

being dominated by the Indian Fijians (Iyer, 2007, 132). Nevertheless, the divide in Fijian

society exists in a lot of different ways, including economic and political, but the main divide

is a racial one, with the two groups having different customs, belonging to different tribes, CEU eTD Collection and having different phenotypical features. This means that Fiji has race as a main divide.

When Lijphart argued that India was a consociational democracy, he claimed that,

while India has a majority Hindu population, the Hindus are so divided by language, cast, and

sect, that they do not form a political majority (Lijphart, 1996, 261). However, that is exactly 21

what happened, in 2014 the Indian people elected the BJP party (Burke, 2014). The BJP is a

Hindu nationalist party, which means that it believes that, first and foremost, India is a Hindu

nation (Seshia, 1998, 1036). While Lijphart was right, there are several different minorities in

India regarding caste and language, the main divide in Indian society is a religious one. This

is even more exemplary by the large amount of violence especially between the Hindu

community and the Muslim community, an example of which is the 2002 Gujarat pogroms,

which was started with the killing of 58 Hindus and resulted in the widespread murdering of

members of the Muslim community in the state of Gujarat (Bilgrami, 2013, 143). This shows

that, while there is a lot of diversity in India, the main divide is religious one, not a racial one.

Israel has historically been an immigrant country, and therefore it is not surprise that

the country is racially diverse (Phinney, et al., 2001, 500). Outside of the Jewish population

there is the Palestinian population, who have Israeli citizenship but are mainly Muslim or

Christian (Kook, 2017, 2046). The divide between the Jewish and Palestinian population is

also a language divide, with the Palestinians speaking Arabic and going to Arabic language

schools (Kook, 2017, 2046). These divides are also played out in the political arena, there is a

Palestinian party and a Jewish ultra-orthodox party both being represented in the Israeli

parliament (Kook, 2017, 2045). However, while the divide is one that can have racial

components, the divide is mainly between the Jewish population and the non-Jewish

population, and while there are differences between the groups, a religion is not a race.

Therefore, we will not consider Israel to have a racial divide.

Kosovo’s divide is similar to the divide in Bosnia. The society consists mainly of CEU eTD Collection Albanians, but also has a Serbian group (Taylor, 2005, 440). The divide in Kosovar society

mainly stems from the fact that the Albanians want an independent Kosovo, while the

Serbians want Kosovo to remain a part of Serbia (Taylor, 2005, 440). This divide has led to

violent conflict in the past as well, with the Kosovo independence war only ending after 22

NATO bombed the Serbian government (Jenne, 2009, 281). However, while the divide in

Kosovo is one that has led to violence, it is not a racial divide, but one based on nationality.

The main divide in Lebanese society is religious divide, to the extend where the

political system is divided up by religion as well (Dekmejian, 1978, 254). Certain post in

Lebanon are reserved for certain religious groups, the Maronites hold the presidency, the

Sunnis hold the premiership, the Shi’ites hold the Chamber Speakership (Dekmejian, 1978,

254). While the divide in Lebanese society has led to violence at some points, it is mainly a

religious divide, not a racial one.

Like many European countries, the divide in Luxembourg society is based on

language (Magone, 2016, 97). There are three major languages in Luxembourg, French for

the public administration, German which is one of the main languages in the country, and

Luxembourgish which is considered the native language of Luxembourg (Magone, 2016, 97).

Luxembourg also has a large population of foreign nationals, mainly from Portugal, Italy, or

former Yugoslavia (Magone, 2016, 97). This however, is not a racial divide, as in this case,

the languages are not part of a greater racial conflict. Therefore, we will not consider

Luxembourg to have a racial divide.

The divide in Malaysian society is mainly a racial one, between the indigenous Malay

and the Chinese (Singh, 2001, 45-46). While there are several subcategories within the two

groups, for instance the Chinese can be Cantonese, Hokkien, or Kheks, and the indigenous

Malay can be Javanese, Jakun, or Banjarese, the two main racial groups are indigenous Malay

and Chinese (Singh, 2001, 46). Even Rabushka and Shepsle mention Malaysia as having a CEU eTD Collection racial divide in their book, where they mention that, while indigenous Malay and Chinese

belong to the same Mongoloid race, they are subcategory of that race, and therefore there is

an racial division (Rabushka and Shepsle, 1972, 8). This research would agree with that

because the Indigenous Malay and Chinese Malay do have different phenotypical features,

and therefore the divide in Malaysian society will be classified as a racial divide.

The two largest national groups in Macedonia are the Macedonians and the Albanians

(Staniševski and Miller, 2009, 557). While there are some smaller groups or Turks and

Roma, most the population belongs to the Macedonian or Albanian group (Staniševski and

Miller, 2009, 557). The divide between the two groups has caused some conflict in the past,

for example, the Albanians protesting the government’s decision to not allow the Albanians to

fly their flag on public buildings during the holidays next to the Macedonian flag, these

protests have ended in riots (Staniševski and Miller, 2009, 558). However, while the

Albanians and Macedonians might be two different national groups, they are from the same

racial group when it comes to our definition of race, therefore Macedonia will not be

considered to have a racial divide.

When it comes to cleavages in Dutch society, the main divides are political and

religious. The Netherlands has historically had a fragmented party system; however, the

fragmentation seems to have grown (De Sio, Paparo, 2018, 53). The Freedom party of Geert

Wilders is openly anti-Muslim, and has even been powerful enough to support a minority

coalition (Marzouki, McDonnell, Roy, 2016, 67-74). This has created a cleavage in the Dutch

party system between the more cosmopolitan parties and the more nationalist parties, with the

cosmopolitan parties focusing more on the environment, and the nationalist parties focusing

more on the national identity (De Sio, Paparo, 2018, 55). The Netherlands also has a religious

divide, with a substantial Protestant, Catholic, and Muslim community (Schmeets, 2016, 5). CEU eTD Collection However, while the Netherlands is racial diverse, with about twenty-two percent of the

population having an immigrant background, the main divide in Dutch society is a political

one (CBS, 2017). Therefore, the Netherlands does not have an racial divide.

Nigeria has many different tribes living within its borders (Jinadu, 1985, 74). There is

the Igbo population, the Edo population, and the Ijaw population (Jinadu, 1985, 74). The main

divide is clearly tribal based, belonging to a different tribe meant having a better position in

society (Jinadu, 1985, 73). The Igbo, Yoruba, and Hausa-Fulani, for instance, long had a

hegemony on the political, social, and economic life in the country (Jinadu, 1985, 73).

However, while the tribes are different, they do belong to the same racial group, as they do

not have different phenotypical features. Therefore, we will not consider Nigeria to have a

racial divide.

At the core of the conflict in Northern Ireland is a divide between nationalities.

Northern Ireland is divided up in two nationalities, the British and the Irish, with the British

being the majority (Tonge, 2002, 1). The difference between the British and the Irish is found

in different cleavages. Most the people in Northern Ireland are Protestant, and many of the

Protestants are British, less than fifty percent of the population is Catholic, and many of the

Catholics consider themselves to be Irish, creating a religious cleavage as well (Tonge, 2002,

1-2). The cleavages are also visible in the political arena, with many of the British favoring

the remaining of Northern Ireland within the United Kingdom, they are known as Unionists,

and most the Irish population favoring a return of Northern Ireland to Ireland, known as

Nationalist (Tonge, 2002, 1). However, while there are many different divides in Northern

Ireland, the British and the Irish do not have different phenotypical features, therefore, the

divide in the society of Northern Ireland will not be considered a racial divide.

South Africa is perhaps the best example of a society that has a racial divide. South CEU eTD Collection Africa is a multiracial country, which includes black, Indian, white, and mixed South Africans

(Lijphart, 1985, 3). The racial diversity in South Africa has led to the creation of Apartheid in

the past, which was a form of segregation created by the white South Africans to oppress the

non-white South Africans (van der Vyver, 1991, 745-746). The system of Apartheid became 25

official state-policy in 1948 (van der Vyver, 1991, 745). In 1994 Apartheid officially ended

with the passing of a new constitution, but that does not mean that the racial divide in society

ended suddenly (Lijphart, 1998, 144). The different racial groups have different phenotypical

features and therefore we will consider South Africa to have a racial divide.

Suriname is an racially very diverse society. The society is comprised of East Indians,

Maroons, Creoles, Javanese, and mixed, with a small group of Chinese and native tribes

(Veenendaal, 2019, 6). The first political parties were established on the different racial

groups, and racial differences have led to tensions in the past (Veenendaal, 2019, 7).

Suriname had a coup in the 1980’s, after which a civil war broke out between the Surinamese

military and a Maroon insurgency (Veenendaal, 2019, 7). Therefore, the racial divide is the

main divide in Suriname society, and we will consider Suriname to have a racial divide.

While there are several divides in Switzerland, most of them do not seem to be very

strong divides. There is a religious divide, between the Protestants and the Catholics, the

language divide between different regions of Switzerland, and a political divide between

different parties on the political spectrum (Vatter, 2016, 66-67). However, none of those

divides are an racial divide, therefore Switzerland will not be considered to have a racial

divide.

The divide in Uruguay society is similar to the divide in the Colombian society, in that

it is a political divide. There are two main parties, the Colorado, and the Blanco parties

(Cason, 2002, 92). The divide has led to civil wars which happened periodically until 1904

(Cason, 2002, 92). Uruguay also experimented with a consociational democracy system, but CEU eTD Collection that ended in 1967 (Lijphart, 1969, 213). Nevertheless, the divide in Uruguayan society was

not a racial one, but a political divide.

2.2 Consociational democracy or not?

Next let us look at the political system that the countries with racial conflict have,

these are the following countries; Fiji, Malaysia, South Africa, and Suriname. While these

countries were all once seen as being consociational, but it is important to look at why they

were classified as consociational in the first place. Therefore, we will go over each country

that has an racial divide to look if they at some point had a system that can be classified as

consociational. We will do this by looking at the definition of consociationalism that was

provided above by Lijphart and Bogaards. The results of this are in table 2, and following that

is the explanation as to why these countries are considered consociational and at what time.

Table 2

Fiji is a former colony of the United Kingdom, and became independent in 1970 (Iyer,

2007, 132). The country has had different political systems, but it is commonly accepted that

they were consociational directly after independence in 1970 (Iyer, 2007, 133). The post-

independence constitution implemented important an important power-sharing framework

CEU eTD Collection that gave a large amount of political power to the indigenous-Fijians, to make up for the large

economic power that was held by the Indo-Fijians. The constitution mandated racially

separate representation in the House of Representatives, with twenty-five seats being racially

allocated but voted for by electors of all races, ten for the indigenous-Fijians and Indo-Fijians

and five for Others, and twenty seven seats were distributed to the separate communities,

twelve for the indigenous-Fijians and Indo-Fijians and three for Others (Ghai and Cottrell,

2007, 644-645). Of the twenty-two seats in the Senate eight were nominees by the Great

Council of Chiefs, an indigenous-Fijian institution, seven were appointed by the prime

minister, six by the leader of the opposition, and one by the Council of Rotuma (Ghai and

Cottrell, 2007, 645). Because of the majoritarian election model that came from their former

colonizer, this meant that the House of Representatives would largely be indigenous-Fijian

dominated, but it would require a small amount of support from European, part-Europeans,

and Pacific Islanders, which had traditionally been allies of the indigenous-Fijians (Ghai and

Cottrell, 2007, 645). The constitution also protected traditional indigenous-Fijian institutions,

the Great Council of Chiefs being one of them, and established separate indigenous-Fijian

courts (Ghai and Cottrell, 2007, 645). The seats in the House of Representatives that were not

given to any race meant that parties had to appeal to all groups, which gave an incentive for

racial groups to not just listen to their own group but to also find inter-segmental agreements.

This constitution shows clear signs of consociationalism, as it had mandatory representation

of all groups, but also incentives for inter-segmental agreements, and protection of the rights

of different groups to have their own system, the protection of indigenous-Fijian institutions

being an example of that. This means that we can consider Fiji to have been consociational.

Malaysia is another former British colony, and gained independence in 1957 (Sani,

2009, 98). The divide in Malaysian society is mainly one between indigenous-Malays and

Chinese-Malays, with the indigenous-Malays being afraid of being dominated by the Chinese- CEU eTD Collection Malays (Sani, 2009, 99). Malaysia is assumed to have had a consociational system between

1955-1969 by Lijphart (Lijphart, 1979, 512). This is in part to the grand coalition known as

the Alliance, which had UMNO for the Malays, the MCA for the Chinese, and the MIC for

Indians (Haque, 2003, 246). The Alliance was first elected in 1955, and formed the 28

government in 1957 (Haque, 2003, 246). The consociational system also came with a

constitution that allowed for “special rights” for indigenous-Malays, which included special

rights in education, business, and the public service (Haque, 2003, 244). Malaysia also had an

informal veto, it was impossible to change the constitution without a two thirds majority,

meaning that the parties had to accommodate each other for them to change the constitution,

as no party in the Alliance held a two thirds majority (Haque, 2003, 246-247). With this we

can find that Lijphart was right in his classification of Malaysia as a consociational system, as

Malaysia incorporated different consociational practices in their constitution and political

system in general.

South Africa is a historically divided society, and the history of Apartheid divided the

society even further, with divisions being along racial lines, which divided the society up into

four groups, white, colored, Indian, and black African (Traniello, 2008, 28-30). With the end

of Apartheid also came a call for a different type of political system, as the Apartheid system

had excluded most of the society, the call for a consensus-based system came from the need

to avoid uncertainty and volatility (Traniello, 2008, 35-36). In 1994, a new system was

adopted, which involved the passing of a new interim constitution (Lijphart, 1998, 147). The

interim constitution called for a Government of National Unity, which included all parties that

had a minimum of five percent of the seats in the National Assembly (Lijphart, 1998, 146).

The Government of National Unity included several different parties, including the National

Party, which represented the white population and the ANC which represented the black

African and Indian population (Lijphart, 1998, 147-148). The constitution also guaranteed a CEU eTD Collection right for people to establish an educational institution based on things such as culture,

language, or religion, as long as there was no discrimination based on race (Lijphart, 1998,

146). Elections were also done by proportional representation, and a two thirds majority was

necessary for amending the constitution (Lijphart, 1998, 146). The 1994 constitution of South 29

Africa was a consocational constitution, it included all four principles of consociationalism,

even if some were stronger than others, and therefore South Africa can be considered

consociational, at least for the time that it had the interim constitution.

Suriname is a former colony of the Netherlands (Singh, 2014, 133). The country

became independent in 1975, and adopted the consociational model of its former colonizer,

the Netherlands (Veenendaal, 2019, 6). The country is comprised of multiple racial groups,

East Indians, Maroons, Creoles, Javanese, who are from the island of Java in Indonesia, and

Chinese, there is also a part of the population that is mixed (Veenendaal, 2019, 6). Suriname

uses a proportional representation electoral system, and the country also has multiple parties

which represent the different racial groups (Veenendaal, 2019, 6-7). For a long time,

Suriname had a grand coalition, which consisted of the parties of different racial groups and

was known as the Front for Democracy and Development (Veenendaal, 2019, 7). Suriname

also has different electoral districts, which gives a small amount of segmental autonomy to

the people in each district, as they can chose their own representatives (Veenendaal, 2019, 6).

These all show that Suriname has been consociational, at least post-independence from the

Netherlands.

CEU eTD Collection

Chapter 3

3.1 The end of consociationalism?

Let us now look at the countries that had a consociational system, but stopped being

consociational. What made these countries stop being consociational? Specifically, what were

the defining factors in the ending of their consociational democracy? The countries that have

had a consociational system but stopped being consociational are the following; Fiji,

Malaysia, South Africa, and Suriname.

Before independence Fiji already had an election system that consisted of separate

rolls for indigenous Fijians, Indian Fijians, and a third roll for voters who were not a part of

those two groups (Iyer, 2007, 132). The separate roll system was supported by the chiefs of

the indigenous Fijians; however, the Indian Fijians did not support this system because it

seriously disadvantaged them (Iyer, 2007, 132). For a while after independence the system did

work, that is until 1987, when an Indian dominated government was elected (Iyer, 2007,133).

Between 1970 and 1987, Fijian politics was mainly ruled by the Alliance Party (Reddy, 2011,

195). The Alliance Party was an inter-racial party which had organizations from all different

backgrounds, including the Indo-Fijians, who were willing to have indigenous-Fijian

dominance in the political sphere in exchange for professional and educational advancement

without being disturbed by politics (Alley, 2000, 516). However, while the Alliance Party

received a larger percentage of the Indo-Fijian vote for the seats in the House of

Representatives that were not racially allotted, larger than the Indo-Fijian party NFP, it also

had to maintain the support of the indigenous-Fijians (Ghai and Cottrell, 2007, 647). This was CEU eTD Collection especially difficult because of the existence of militant indigenous-Fijian parties which were a

serious contender for the indigenous-Fijian vote (Ghai and Cottrell, 2007, 647). The Alliance

Party started appealing to the indigenous-Fijian population more, which meant disregarding

the Indo-Fijian population more (Ghai and Cottrell, 2007, 647). By 1977 the Alliance party

had lost most of the Indo-Fijian votes, but was still able to win because of the Other group in

the electoral system, which had historically sided with the indigenous-Fijians (Ghai and

Cottrell, 2007, 647-648). But in 1985 the Fijian Labour Party was formed, which focused not

on race but instead on class, but was still largely supported by the Indo-Fijians (Ghai and

Cottrell, 2007, 648) (Alley, 2000, 516) (Fraenkel, Firth, and Lal, 2009, 3). The Fijian Labour

Party quickly won support amongst the indigenous-Fijians as well, who felt left behind by the

Alliance Party which they saw as corrupt and unaccountable (Alley, 2000, 516). With the new

elections in 1987, the Fijian Labour Party formed a coalition with the Indo-Fijian NFP,

creating a government that was Indo-Fijian dominant (Ghai and Cottrell, 2007, 649).

However, the new government did not use this power to fully create an Indo-Fijian

government, instead it divided up the post of cabinet ministers evenly among the racial

groups, with seven cabinet ministers being indigenous-Fijian, and seven being Indo-Fijian,

with the Indo-Fijian cabinet ministers holding portfolios that had typically gone to Indo-

Fijians (Ghai and Cottrell, 2007, 649).

The Indo-Fijian dominant government created unrest among the indigenous-Fijian

community, which was exploited by army officer Sitiveni Rabuka (Iyer, 2007, 133). Sitiveni

Rabuka mounted a bloodless coup that ended the democratic system, and declared Fiji a

republic with a military government (Alley, 2000, 516). The military government lasted for

three years, after which Rabuka ordered the formation of a new democratic government with

a new constitution that was incredibly discriminative against the Indo-Fijian population, CEU eTD Collection effectively barring them from holding important government positions such as Prime

Minister, and only giving them twenty-seven out of seventy seats in the House of

Representatives (Alley, 2000, 516). Fiji never fully went back to the system of government

that it held between 1970-1987, and therefore it is largely considered to only have been

consociational between 1970-1987.

There were two main factors that led to the 1987 coup. First and foremost was the

winning of the election by the Fijian Labour Party and the NFP, which led to an Indian

dominated government (Alley, 2000, 516). Another factor was the distrust of the indigenous-

Fijian elites, specifically the Alliance party, which resulted in the election of the NFP-Fijian

Labour Party Coalition (Alley, 2000, 516) This distrust of the elites was also used by the coup

perpetrators, who claimed to have staged the coup in the name of the Great Council of Chiefs

(Fraenkel, Firth, and Lal, 2009, 33). These were of course not the only factors that started the

coup, but they were the most important ones.

Fiji’s experiment with consociationalism lasted seventeen years, but failed because of

conflict between the Indo Fijian group and the indigenous Fijian group. The Indigenous Fijian

elites lost the control over their population, which was used by the military to stage a coup.

Here we see two perfect examples of what Barry claimed was the problem in racially divided

societies. The first being the fact that racial groups do not have a set way of interpreting the

world, unlike religion which can follow rules (Barry, 1975, 502). This can explain why the

indigenous elites lost control, whereas a religious leader can fall back on the word of God or

some religious dogma, the indigenous leaders had no way of explaining their actions as being

for the good of the people, instead their authority was questioned easily by those claiming to

better speak for the people. The second is Barry’s claim that religious and class groups are

about organizations, whereas racial groups are not necessarily organized, you do not need an CEU eTD Collection organizational structure to start a riot just a way of recognizing ‘the other’ which race makes

possible (Barry, 1975, 502). This is exactly what happened in the 1987 coup, even though the

military had no real way of organizing the indigenous population, they were able to gain

control, they did not need to claim religious authority, instead he just had to claim they did it 33

for the good of the indigenous population. Another factor that also comes into play is that of

the vertical and horizontal ethnic differentiation as mentioned by Horowitz (Horowitz, 1971,

232). The Fijian society has a clear vertical ethnic differentiation, meaning that the divide is

hierarchical based (Horowitz, 1971, 232). The Indo-Fijians held more land and therefore a

larger part of the economy, this created a hierarchical system where the Indo-Fijians had more

influence than the indigenous-Fijians. This was meant to be compensated by the larger

influence of the indigenous-Fijians in the political sphere, creating a horizontal ethnic

differentiation, but when that influence fell away with the winning of the elections by the

Indo-Fijian parties, the society went back to a vertical ethnic differentiation, something which

the indigenous-Fijians did not agree with. The vertical and horizontal ethnic differentiation is

something that was noticeable in other countries as well, something which will also be further

discussed in the conclusion.

Malaysia has a variety of racial groups, but the Chinese-Malay and indigenous-Malay

were the two main racial groups fighting for power in Malaysia (Singh, 2001, 45). The

tension between the two racial groups existed in Malaysia before its independence from the

United Kingdom, and because of this power sharing was an important precondition for

independence (Singh, 2001, 48). The bargain that was made for independence was a simple

one, indigenous-Malays received special political power, in exchange for the citizenship

rights for non-Malays (Singh, 2001, 45). This was mostly done because of the economic

power of the Chinese-Malays, the economy was mainly a Chinese domain, although there was CEU eTD Collection of course a difference between the Chinese-Malay elites and the rest of the Chinese-Malay as

well (Singh, 2001, 45-50). This meant that the indigenous-Malay were more powerful

politically, while the Chinese-Malay were more powerful economically (Singh, 2001, 45).

The political system function relatively well, while there were still tensions between the 34

communities, the consociational system forced parties to incorporate inter-racial politics as

well, as it was the only way for them to compete with the interracial Alliance that led the

country (Singh, 2001, 49). However, the ending of consociationalism in Malaysia started with

just that Alliance. In 1969, the Alliance lost heavily in the elections, not getting a two-thirds

majority in the parliament (Singh, 2001, 49). The reason that the Alliance lost in the election

is directly related to the control over the political sphere that the indigenous-Malay had. The

government past several laws that favored the Malay language and culture and favored

indigenous-Malay in the economic sphere as well, for example creation of the Bank

Bumiputra that gave credit and services only to indigenous-Malay, and making the Malay

language the official language of Malaysia (Klitgaard and Katz, 1983, 335-336). This

preferential treatment angered the Chinese-Malay, which they showed by no longer voting for

the Malaysian Chinese Association, which was part of the Alliance (Klitgaard and Katz, 1983,

336). At the same time, the indigenous-Malay were also angry with the Alliance, while the

Alliance heavily favored the indigenous-Malay, they were still economically and

educationally behind the Chinese, which they saw as the Alliance’s policies not working fast

enough (Klitgaard and Katz, 1983, 336). The indigenous-Malay also chose parties outside of

the Alliance, and this loss of votes led to the Alliance losing the majority in parliament after

the 1969 election (Klitgaard and Katz, 1983, 336).

The loss of the majority by the Alliance meant an end to the Malay dominance in

Malaysian politics (Singh, 2001, 49-50). This loss was the start for riots that took place on

May 13, 1969, in which hundreds of Chinese-Malay were killed (Singh, 2001, 50). The loss CEU eTD Collection of political power by the indigenous-Malay meant that the delicate power structure, of

economic power belonging to the Chinese-Malay and political power belonging to the

indigenous-Malay, was officially off-balance, and the indigenous-Malay responded with

violent riots (Singh, 2001, 50). The response to the riots was an attempt to restore that power 35

sharing structure, but instead it ended up giving most power to the indigenous-Malays (Singh,

2001, 50). The riots were used to push forward a new state-building strategy that focused

primarily on the indigenous-Malay, with the promotion of Malay culture, religion, language,

and the advancement of indigenous-Malay in the economic sphere (Singh, 2001, 50). The

New Economic Policy, or NEP, as it was called, turned over a large proportion of the

products of new economic growth to the Bank Bumiputra, which in turn gave it to the

indigenous-Malay community (Klitgaard and Katz, 1983, 337). Malaysian universities also

started adopting Malay as its official language, previously it had been English, new

universities were primarily founded for the indigenous-Malay, while the Chinese-Malay

community was denied the right to found their own university (Klitgaard and Katz, 1983,

337). These are just some of the examples of the way that the country changed in favor of the

indigenous Malay after the riots of 1969, but what is clear is that democracy in Malaysia

slowly became smaller, and was replaced by a group control by the indigenous-Malay over

the Chinese-Malay, effectively ending consociationalism in Malaysia (Singh, 2001, 50).

The case of Malaysia shows clear signs of what Barry saw as the problem with

consociationalism in racially divided societies. The ability to recognize the different groups is

one that is very clear, the 1969 riots ended up killing hundreds of Chinese-Malay, something

which would have probably been harder to do if they did not have physical features that are

different from the indigenous-Malay, which relates back to Barry’s first reason for problems

with consociationalism in racially divided societies, the ability to recognize the other groups

quickly (Barry, 1975, 502). The argument that race is about solidarity groups, not CEU eTD Collection organizations, is also visible in the case of Malaysia (Barry, 1975, 502). The difference of

wealth between the Chinese-Malay elite and the rest of the Chinese-Malay did not matter to

those who participated in the riots, instead anyone who belonged to the Chinese-Malay group

was targeted, just as the indigenous-Malay did not need a calling from their elites to start the 36

riots, they just needed enough people to join. The fourth reason Barry proposed, that racial

conflict is not about how a country should be run but if it should be a country at all and who

belongs to that country is also visible in Malaysia (Barry, 1975, 502). The racial conflict in

Malaysia goes back to that exact question, if the Chinese-Malay are even Malay, this is

especially visible in the bargain that was made to give the Chinese-Malay citizenship, it

required the indigenous-Malay to receive large preferential rights (Singh, 2001, 45). These

three reasons all attribute to the downfall of consociationalism in Malaysia, and the taking

over by the indigenous-Malay community. It is very clear that the loss of political power was

an important factor to the ending of consociationalism in Malaysia, but the inequality in the

economic sphere between the indigenous-Malay and Chinese-Malay should also not be

underestimated. This is similar to the case of Fiji as well. Just as in Fiji, Malaysia was a

society that had a vertical ethnic differentiation that was compensated by the consociational

system which attempted to turn it into a horizontal ethnic differentiation system. However, as

in the case of Fiji, this failed by the loss of elections, which made the society turn back into a

vertical ethnic differentiation system. The case of Malaysia is a perfect case for Barry’s

argument, but is also very similar to the other cases studied in this research. The link between

this case and the others will be further discussed in the conclusion.

South Africa’s 1994 constitution was never meant to be one that lasted forever, instead

the constitution was an interim constitution, meant to last until a new, final constitution, was

made (Hart, 2003, 7-8). However, during the interim constitution was used for an election,

specifically the election of 1994, which elected the parliament that would create a new CEU eTD Collection constitution (Hart, 2003, 7-8). The 1994 constitution mandated a government of national unity

which included all parties, and a government of national unity was created after the 1994

election (Lijphart, 1998, 147). However, while the 1994 constitution was largely

consociational, the 1996 constitution was not. The 1996 constitution no longer mandated a 37

government of national unity (Lijphart, 1998, 147). This paved the way for the ANC party to

win the majority in multiple elections, making South Africa a de facto one party state

(Campbell, 2014). But why did the new constitution no longer mandate a government of

national unity? It is of course impossible to know what exactly went on during the creation of

the new constitution, but the fact that the National Party left the coalition government might

have something to do with it (Lijphart, 1998, 147). The National Party was a party that

represented the white community of South Africa, a minority in South Africa (Lijphart, 1998,

148). The National Party left the coalition government in 1996, the year that the new

constitution was adopted, so whether they left because a government of national unity was no

longer necessary or if they left before that is not clear, however what is clear is that the

National Party would have never gotten into government if not for the mandating of a

government of national unity (Lijphart, 1998, 147). The National Party represented the

minority white population, and with the new election system of proportional representation,

there was no possibility that the National Party was ever going to win a majority again

(Lijphart, 1998, 147-149). The ending of the unity government meant an ending to the

consociational system in South Africa, as it no longer had coalition governments anymore. In

his paper on this, Lijphart argues that South Africa does still have a government of national

unity in a way, as the ANC is a strongly multi-racial party that represents and he has

representatives from all the major racial groups in South Africa (Lijphart, 1998, 148).

However, while this might be true, the ANC is still just one party, and differences of opinion

between the party and a party member in government are generally not accepted. The CEU eTD Collection constitution allows for members of parliament to be forced to leave parliament when they are

forced out of the party or leave the party themselves (Mattes, 2002, 24). While this is of

course just a part of the constitution, and it does not necessarily have to be followed, the ANC

has expelled people from its party, and in that way from parliament, in the past (Mattes, 2002, 38

25). Besides that the ANC has also gotten more centralized, candidates for local elections,

such as mayoral races and provincial premiership, are elected through a central committee,

meaning that even though South Africa is largely decentralized according to the constitution,

the ANC has closed that decentralization by being heavily involved in local elections (Mattes,

2002, 25). The ANC has also allowed for “crossing the floor” meaning the switching of

alliance between parties, most noticeably the Democratic Alliance party joined the ANC in a

coalition in the city government of Cape Town, giving the ANC control of the last major city

it did not already control (Mattes, 2002, 26). These things show that, while the ANC might be

racially diverse, it is not ideologically diverse, meaning that opposition within the ANC is not

allowed. This would mean that, even if the ANC would represent different groups within the

South African society, a group that would disagree with the ANC’s policies would have no

way of stopping it. A good example of this is the land reform that is being discussed in South

Africa right now (Toyana, 2019). South Africa has historically had unequal land distribution,

72 percent of the land is in the hand of white farmers, whereas white people are less than ten

percent of the South African population (Clark, 2019). While land reform is a controversial

issue, it is clear that the white farmers have no way of being represented in this issue. The

ANC has spoken out for land reform, and even made it a key issue in the May 2019 election,

meaning that even if the white farmers are part of the ANC, they have no way of voicing their

opinion within the party (Clark, 2019). The ending of consociationalism in South Africa was

not different from the other countries that were studied, mainly that it was a rather peaceful

process. The changing of the constitution and the taking over of the government by the ANC CEU eTD Collection was not done through a coup or riots, but instead through the process of elections and

peaceful debate. However, the rise to power of the ANC is an interesting one, it effectively

took over any opposition possible. Relating it back to Barry is difficult, because the ending of

consociationalism in South Africa did not go as Barry predicted consociationalism to fail in 39

racially divided societies. However, the fourth reason that Barry gives about why

consociationalism does not work in racially divided societies might work best here. Barry

mentions that racial conflict, unlike religion or class conflict, is about values, it is not about

how a country is run, but if the country should even exist at all, and if it should, who belongs

in that country (Barry, 1975, 503). This is something that can be seen in the land reform issue

in South Africa, the question as to who belongs to the land or who owns the land, in this case

a very physical representation of the country, is something that is disputed. While this goes

back to a legacy of Apartheid that created an uneven ownership of the land in the first place, it

is also about who belongs to the country (Clark, 2019). The ending of consociationalism in

South Africa was clear when the ANC became the biggest party, but that does not mean that

South Africa is no longer a functioning democracy, just not a consociational one. South

Africa, however, is an outlier, it is different from the other countries studied. This also shows

in the way that the country dealt with its vertical ethnic differentiation and horizontal ethnic

differentiation. It is clear that, before the ending of Apartheid, South Africa had a system of

vertical ethnic differentiation, and while this did not fully turn into a system of horizontal

ethnic differentiation, the great power of the ANC did compensate a little for the economic

power of the white minority. Perhaps this is also an important factor as to why South Africa is

such an outlier case, the differences in size between the two racial groups is much larger in

South Africa, meaning that the loss of political power by the group that is economically less

powerful, the black South Africans, has yet to happen, which is unlike the other countries that

were studied so far. CEU eTD Collection Similar to many former colonies, Suriname adopted the political system of their

former colonizer, which in the case of Suriname meant the consociational system of the

Netherlands (Marchand, 2014, 345-347). However, the adoption of this political system did

not take place after independence, Suriname was still under the Dutch crown and the 40

Netherlands was still responsible for defense and international relations when, in 1946,

Suriname became a democracy (Marchand, 2014, 345-346). Suriname did not fully become

independent from the Netherlands until 1975, and it was shortly after that that the

consociational system ended in Suriname (Marchand, 2014, 347).

The events that led to the ending of consociationalism in Suriname are like those of

other countries that are used in this research. Suriname has several racial groups, but the most

influential ones have usually been the Creole racial group, which are descendent of former

slaves. The Creole group was allowed to vote in 1901, before that the only group being

allowed to vote were the Dutch elite (Marchand, 2014, 345). Even when Suriname became

semi-independent the Creole parties still had large amounts of power, but the country was

consociational (Marchand, 2014, 346). The largest disagreement happened between the

Creoles and the Hindustanis, where the Creoles favored independence from the Netherlands,

the Hindustanis did not (Marchand, 2014, 346). The Hindustanis feared that if Suriname

became independent the Creoles would become dominant, and they feared losing their

preferential agricultural rights that were given to them by the Netherlands (Marchand, 2014,

346). In 1973 the Creole party, NPS, took over, they had previously worked in an alliance

with the Hindustani party, but this time they won a majority (Marchand, 2014, 346). Soon the

NPS called for independence, and the country officially became independent from the

Netherlands in 1975 (Marchand, 2014, 346-347). The newly independent democracy did not

remain a democracy for long, in February 1980 the military overthrew the government, after

being angry at the economic situation and the continuous influence of the Netherlands CEU eTD Collection (Marchand, 2014, 347). Sergeant Desi Bouterse, a member of the Creole racial group, was the

leader of the coup, and declared martial law, and the military period started in Suriname

(Marchand, 2014, 347). During the military period, Suriname suffered from the withdrawal of

developmental aid from the Netherlands, which hit the economy hard, but it also made the 41

economy more informal, creating a large black market in the country and the country became

a part of the international cocaine trade (Marchand, 2014, 347). The military rule formally

ended in 1987, with the first democratic elections, but that does not mean that democracy was

fully restored in Suriname (Marchand, 2014, 347). Like the case of Fiji, the perpetrators of the

coup were never punished, in the case of Suriname, Desi Bouterse was still in politics, albeit

with a small political party, but his stronghold on the nation persisted through his position in

the military (Marchand, 2014, 347). This was especially visible in December 1990, when

Bouterse called the cabinet and said that the then president, Shankar, who was democratically

elected, could no longer govern, and pushed the Jules Wijdenbosch, who was a member of the

NDP, Bouterse’s party (Marchand, 2014, 347).

Consociationalism fully ended in Suriname with the 1980 coup, even though

democracy was restored after the military rule, whether fully or not is up to debate, Suriname

never went back to being a consociational democracy again. But, while the coup was the

ending of consociationalism, the move towards the ending of consociationalism was started

before that. The NPS becoming the biggest party started the ending of coalition governments

in Suriname, and with the call for independence also came a division between the different

racial groups. When looking back at Barry’s criticism of consociationalism in racially diverse

societies, the last two reasons seem to be the most applicable for Suriname. racial groups do

not have a set way of interpreting the world, and racial conflict is not about how a country is

run, but if it should even be a country at all (Barry, 1975, 502-503). The division between the

Hindustani party and the Creole party was about the independence of Suriname, a question CEU eTD Collection that was not about how to run the country, but how the country should exist in the first place.

The lack of interpretation of the world in racial groups is mostly visible in the first post-coup

election, in which the NPS, the Creole party, won more seats than the NDP, the party that

Desi Bouterse belonged to (Marchand, 2014, 347). What the population wanted and what the 42

Creole elite wanted were two different things, Bouterse wanted power, but the Creole

population was not ready to give it to him in a democratic election. The fact that he forced the

government to make a member of his party president also shows a disconnect between the

elite and the people, and a definite lack of how to interpret what was best for the country.

Bouterse is currently president of Suriname, with his NDP party winning the majority of the

seats in the last election (Kuipers, 2015). While Bouterse has committed many crimes during

his time as military leader, including trafficking drugs and the killing of Maroons in a six-year

internal war, he does not seem to have been punished for that in his ability to be elected

(Marchand, 2014, 347-348). The ending of consociationalism in Suriname was not a peaceful

one, but instead it was the violent takeover by the Creole elite of the country’s government,

while this might not have been fully supported by the Creole people themselves, it does not

mean that the taking over of one group in the country did not take place. This is similar to

what happened in the other countries studied as well, and happened for similar reasons as

well. What is interesting is that the system of horizontal ethnic differentiation existed in

Suriname, but before they received full independence from the Netherlands. The Creoles were

a favored group, as in they got voting rights before other racial groups except for the

colonizing group, they were evened out by the preferential land rights that the Netherlands

gave to the Hindustani group. With the independence from the Netherlands in Suriname also

came the possibility by the Creole group to take over, and to create a system of vertical ethnic

differentiation. In this sense, it was not a lost election, but still a large political change like the

other countries studied that ended the consociational system, and created a system of vertical CEU eTD Collection ethnic differentiation in Surinamese society.

3.4 Conclusion

"Give me an example, of a multi-ethnic or multicultural society, where the original

population are still living as well. (...) And where there are peaceful community relations. I’m

not aware of any.” (Zembla International, 2018). In the summer of 2018, the Dutch minister of

foreign affairs, Stef Blok, made a speech in front of Dutch citizens who work for international

organizations, in it he claimed that multicultural societies do not work, and that this is because

people do not like to be around those who are unfamiliar to them (Zembla International, 2018).

The statement caused much controversy, but did not lead to the end of Blok’s political career (Ast

and Keultjes, 2018). But was Blok necessarily wrong? If we found anything by looking at these

cases is that, while most of these societies might not experience violent conflict right now, they

have gone through violent conflicts and coups in the past. But that does not mean that Blok is right

in his statement, after all Blok discusses multiethnic societies not multiracial societies. Just as was

mentioned earlier in this research, the difference between ethnicity and race might seem small, but

they do constitute two entirely different concepts. If this was a fact checking research we would

find his statement to only be half true.

But before we go further with the conclusion, let us first make clear what the purpose of

this research is. Some might look at this research and think of these societies as inherently violent

because of the races that live in them. This research strongly condemns those thought, and would

like to state that the analysis of these countries is not done from a primordalism standpoint, but a CEU eTD Collection constructivism standpoint. These societies had a long history of issues such as colonialism,

discrimination, and structural inequality, that all influenced what ended up happening in these

countries. To look at these countries and find that their main problem is race, would be wrong, and

show a limited scope when it comes to problems in this world.

The countries that were researched all had a strong racial divide that played out not just in

the political sphere, but also in the economic sphere. The white minority in South Africa holds

most of the land, which is like the situation in Fiji with the Indian-Fijians holding most of the land

as well and similar to the situation in Suriname with the Hindustani population having inherited

preferential land rights from the Netherlands. In Malaysia, it is not as much about land, however

the Chinese-Malay were in control of most of the economy.

This brings us back to the original criticism of consociationalism, and specifically

Lijphart’s response to it in his book on South Africa. Lijphart writes about consociationalism as

though it is only about the sharing of power, but he is forgetting the emotional aspect of that

sharing. In his book, Lijphart argues against Barry’s statements, saying that Barry’s arguments are

all about elite control, forgetting that reaching inter-segmental agreements are more important.

Lijphart then states that reaching inter-segmental agreements are easier in ethnically divided

societies because, unlike religion or class, ethnic divisions are about emotional incompatibility, not

about logical incompatibility, and therefore more likely to find a compromise (Lijphart, 1985, 96).

But this is exactly what was found in the countries studied. The race riots in Malaysia, the leaving

of the coalition by the National Party in South Africa, and the coups in Fiji and Suriname were all

about emotional incompatibility between groups. The indigenous-Fijians were afraid of being

dominated by the Indian-Fijians, the white South Africans saw themselves losing their political

power as a minority, the indigenous-Malay felt that they were being disadvantaged in comparison CEU eTD Collection to the Chinese-Malaysians, and the Creole group in Suriname felt that the influence of the

Netherlands was still too big. These were all cases of emotions, none of them were about different

interpretations of scripture of who holds the true religion, it was about feeling disadvantaged. In

this sense Lijphart was wrong, emotions are harder to deal with in a political system. 45

Lijphart is right to certain extend in his defense of consociationalism though, although

maybe not in the way he intended it to be. In his book on South Africa, Lijphart argues that ethnic

divides are hard to distinguish from religious divides, giving the examples of Lebanon and

Northern Ireland (Lijphart, 1985, 97). To an extend this is true, even in these cases the racial divide

did not stand alone in bringing a divide to the society. But it was not an intersection with a religious

divide that we see here, but an intersection with a class divide. All four countries studied saw

divides between the racial groups in the economic sector, some larger than others, but also still

large enough to be visible. It is this class divide that was the main cause for conflict in these

countries. This leads us back to the argument of Horowitz as well, the difference between vertical

ethnic differentiation and horizontal ethnic differentiation. All four countries had vertical ethnic

differentiation, with a clear hierarchy between the races, and consociationalism tried to turn these

countries into a system of horizontal ethnic differentiation. This failed when the compensation for

economic power for one group with the political power for another group fell away, turning the

countries, except for South Africa, back into a system of vertical ethnic differentiation. This change

in system can be attributed to the ending of consociationalism, and shows a fault in the

consociational system, as it was possible to have the change in system in first place. Of course this

is not completely the fault of the consociational system, or even consociationalism in itself, but it is

important to remember when it comes to using consociationalism in racially divided societies that

also have an economic divide between the racial groups. A change the power structure can quickly

lead to the end of the consociational system.

But there is another question that comes from this, which is if it really was the racial divide CEU eTD Collection in the societies that led to the ending of consociationalism, or if it is just the system itself that is

doomed to fail? Of course, when looking at these countries, you can see that the racial divide

played an important role in the ending of consociationalism, but perhaps this is something that

happens in any society that adopts the consociational system, and is the fact that these divides were 46

just racial not relevant to the ending of consociationalism, but just the fact that there was a divide at

all. For this we should look back at the countries that were first used, and then the ones that did not

have a racial divide. Of those countries, most of them are democracies that did not have a coup of

race riots, such as Austria, Belgium, Canada, Luxembourg, Netherlands, and Switzerland. For

countries such as Antilles (NL) and Czechoslovakia stopped existing, but this was not because of

the use of violence (Henderson, 1995) (Oostindie, 2006). Bosnia and Northern Ireland did have

violence, but for them consociationalism was implemented to stop that violence, with the Dayton

agreements and the Good Friday agreements (Stroschein, 2014) (Horowitz, 2002). This shows that

for most of the countries it was not the consociational system itself that led to violence, and in

some cases the consociational system helped end the violence. This of course does not mean that

all the countries that had a consociational system did not have violent conflict, but it does show that

it is not solely the fault of the consociational system that three of the four countries studied had

violent conflict, as most countries that were consociational did not have it.

Another important thing that is found in these cases is elite control. As Lijphart said, Barry

focuses a lot of his criticism of consociationalism on elite control, something he sees as being

wrong, but it is something that was seen in the societies that were studied. The loss of elite control,

either through inter elite fighting like Fiji and Suriname, or by losing elections like Malaysia, elite

control is an incredibly important part of the survival of the consociational system. Fiji, Malaysia,

and Suriname, all saw the loss of elite control, which all played out in coups and riots. In the case

of Suriname it was the Creole elite taking back control through the coup, in the cases of Fiji it was

another elite taking over control through the coup, and in the case of Malaysia it was riots that were CEU eTD Collection caused by the loss of control by the elite. By looking at this it is simple to see that elite control is

something that is incredibly important when it comes to consociationalism, and not just the

reaching of inter-segmental agreements as Lijphart argued.

However, there is one case that stands out from the other cases in this study, which is the

case of South Africa. South Africa did not have violent riots or a coup, but instead had a peaceful

transition to democracy, which ended up causing the end of consociationalism, but only because of

the replacement of the constitution. There are two main arguments to be made for why this is. The

first is the fact that none of the other countries had the same history of race relationships as South

Africa did, this makes the case of South Africa deviant from the start. The second argument is that

South Africa never had a loss of elite control, instead it was the elites that caused the peaceful

transition in the first place, through negotiations and elections before the passing of a new

constitution. These two arguments explain why South Africa was a different case, but also make

another argument for elite control, which is important to note when it comes to explaining why

consociationalism fails in racially divided societies.

This research has presented a case for why consociationalism fails in racially divided

societies by showing four racially divided societies that all failed as consociationalist countries,

three of which had the end of consociationalism started by violence. From this we can conclude

that, seeing as consociationalism ended violently in almost all countries that had racial divisions,

consociationalism is incompatible with racially divided societies. However, there are some

important side notes to make from this. First is that all countries did not just have an racial divide,

but also an economic divide, which played out in the racial divide as well. This shows that it is not

just a racial divide that made consociationalism fail, but a class divide that is linked to a racial

divide as well. Out of the four countries studied, the three countries that all had consociationalism

end through violence also had a loss of elite control. This is important because the one country that CEU eTD Collection did not have a loss of elite control did not have a violent end of consociationalism, but instead a

peaceful transition. From this we can conclude that elite control is a very important factor within

consociationalism in racially divided societies.

In the end, there are some lessons to be taken from this, especially when it comes to other

racially divided societies. It is not necessarily consociationalism that fails, but the unequal

divisions between racial groups and the loss of elite control that leads to the failing of

consociationalism. There is also a lesson to be found in the case of South Africa, mainly that of

how to have a peaceful transition in a society. Of course, consociationalism failed in South Africa,

but that does not mean that consociationalism failed. South Africa shows that consociationalism

can lead to a peaceful transition of powers, as long as the elite control is strong. This shows that,

while consociationalism might not work for racially divided societies in the long term, it can work

in the short term, and prevent violent conflict. Further studies on this will be interesting, especially

those that study the use of consociationalism in divided societies when it comes to a peaceful

transition to democracy. However, for this study, the conclusion is clear, consociationalism failed

in racially divided societies, but it was not the system itself that made these societies fail, instead

these societies that had inequalities that made them doomed to fail under almost any political

system, and the loss of elite control made it even worse. Consociationalism is not fully to blame for

the failing, but it also did not help these societies avoid conflict. In this sense the research question

is difficult to answer, consociationalism is not incompatible with racially divided societies, but it

also does not help them, it is like any political system, dependent on the people and the

circumstances that are under it.

CEU eTD Collection

Sources

Adeney, Katharine, “Constitutional Design and the Political Salience Of “Community” Identity in Afghanistan,” Asian Survey, Vol. 48, No. 4, 2008. pp. 535-557.

Akram-Lodhi, Haroon, ”Confronting Fiji Futures,” ANU eView, The Australian National University, 2016.

Alley, Roderic, “The Coup Crisis in Fiji,” Australian Journal of Political Science, Vol. 35, No. 3, November, 2000. pp. 515-521.

Ast, Maarten van, Keultjes, Hanneke, “Minister Blok overleeft motie van wantrouwen na ‘failed state’- opmerking,” AD, 5 September, 2018. https://www.ad.nl/politiek/minister- blok-overleeft-motie-van-wantrouwen-na-failed-state-opmerking~a6dd6ed0/

Barry, Brian, “Political Accommodation and Consocational Democracy,” British Journal of Political Science, Vol. 5, No. 4, October, 1975. pp. 477-505.

Biervliet, W.E., Bovenkerk, F., Köbben, A.J.F., “Surinaamse immigratie: overheidsbeleid en de rol van het sociale onderzoek,” Beleid en Maatschappij, 1975. 337-341.

Bilgrami, Akeel, “Democratic Culture: Historical and Political Essays,” Routledge, 1 February, 2013.

Billiet, Jaak, Maddens, Bart, Frognier, André-Paul, “Does Belgium (still) exist? Differences in political culture between Flemings and Walloons,” West European Politics, Vol. 29, No. 5, 2006. pp. 912-932.

Bingham Powell, Jr. G., “Political Cleavage Structure, Cross-Pressure Processes, and Partisanship: An Empirical Test of the Theory,” American Journal of Political Science, Vol. 20, No. 1, February, 1976. pp. 1-23.

Bogaards, Matthijs, “Consociationalism,” The Wiley-Blackwell Encyclopedia of Social Theory, 2017. pp. 1-2.

Burke, Jason, “Narendra Modi's landslide victory shatters Congress's grip on India,” The Guardian, 16 May, 2014. https://www.theguardian.com/world/2014/may/16/narendra- modi-victory-congress-india-election

Caluwaerts, Didier, Reuchamps, Min, “Combining Federalism with Consociationalism: Is Belgian Consociational Federalism Digging its Own Grave?” CEU eTD Collection Ethnopolitics, Vol. 14, No. 3, 2015. pp. 277-295.

Campbell, John, “South Africa Moving Away From A One-Party State,” Council on Foreign Relations, 9 May, 2014. https://www.cfr.org/blog/south-africa-moving-away-one- party-state

Cannon, Gordon E., “Consociationalism vs. Control: Canada as a Case study,” The Western Political Quarterly, Vol. 35, No. 1, March, 1982. pp. 50-64.

Cason, Jeffrey, “Electoral Reform, Institutional Change, and Party Adaptation in Uruguay,” Latin American Politics and Society, Vol. 44, No. 3, 2002. pp. 89-109.

CBS, “Bevolking; Kerncijfers 2017,” 2017. Accessed 22 May, 2019. https://opendata.cbs.nl/statline/#/CBS/nl/dataset/37296ned/table?dl=107C4

Chandra, Kanchan, “What Is Ethnic Identity And Does It Matter?” Annual Review of Political Science, Vol. 9, 2006. pp 397-424.

Cheng, Jennifer E., “Islamophobia, Muslimophobia or racism? Parliamentary discourses on Islam and Muslims in debates on the minaret ban in Switzerland,” Discourse & Society, Vol. 26, No. 5, 2015. pp. 562-586.

Clark, Christopher, “South Africa Confronts a Legacy of Apartheid,” The Atlantic, 3 May, 2019. https://www.theatlantic.com/international/archive/2019/05/land-reform-south- africa-election/586900/

Clark, Christopher, “South Africa confronts a legacy of Apartheid,” The Atlantic, 3 May, 2019. https://www.theatlantic.com/international/archive/2019/05/land-reform-south- africa-election/586900/

Cordell, Karl, Wolff, Stefan, “Routledge Handbook of Ethnic Conflict,” Routledge, 2011.

Dahlman, Carl, Tuathail, Gearóid Ó., “The legacy of ethnic cleansing: The international community and the returns process in post-Dayton Bosnia-Herzegovina,” Political Geography, Vol. 24, 2005. pp. 569-599.

Dale, Becky, Jeavans, Christine, “India general election 2019: what happened?” BBC, 24 May, 2019. x https://www.bbc.com/news/world-asia-india-48366944

De Sio, Lorenzo, Paparo, Aldo, “The year of challengers? Issues, public opinion, and elections in Western Europe in 2017,” CISE, Rome, 2018.

De Valk, Helga A.G., Esveldt, Ingrid, Henkens, Kène, Liefbroer, Aart C. “Oude en nieuwe allochtonen in Nederland,” Wetenschappelijke Raad voor het Regeringsbeleid, Den Haag, Juli, 2001.

CEU eTD Collection Dekmejian, Richard Hrair, “Consociational Democracy in Crisis: The Case of Lebanon,” Comparative Politics, Vol. 10, No. 2, January, 1978. pp. 251-265.

Deschouwer, Kris, “The Politics of Belgium: Governing a Divided Society,” Macmillan International Higher Education, 3 September, 2012. Second Edition.

Dix, Robert H., “Consociational Democracy,” Comparative Politics, Vol. 12, No. 3, April, 1980. pp. 303-321. 51

Drazen, Petrovic, “Ethnic Cleansing- An Attempt at Methodology,” European Journal of International Law, Vol. 5, 1994. pp. 342-359.

Driessen, Geert, Merry, Michael S. “Islamic Schools in the Netherlands: Expansion or Marginalization?” Interchange, Vol. 37, No. 3, 2006. pp. 201-223.

Dryzek, John S., “Deliberative Democracy in Divided Societies, Alternatives to Agonism and Analgesia,” Political Theory, Vol. 33, No. 2, April 2005. pp. 218-242.

Fox, Jonathan, Rynhold, Jonathan, “A Jewish and Democratic State? Comparing Government Involvement in Religion in Israel with other Democracies,” Totalitarian Movements and Political Religions, Vol. 9, No. 4, 2008. pp. 507-531.

Fraenkel, Jon, Firth, Stewart, Lal, Brij V., “The 2006 Military Takeover in Fiji, A Coup to end all Coups?” ANU E Press, 2009.

Fraenkel, Jon, Grofman, Bernard, “The Failure of the Alternative Vote as a Tool for Ethnic Moderation in Fiji,” Comparative Political Studies, Vol. 39, No. 5, June, 2006. pp. 663-666.

Ghai, Yash, Cottrell, Jill, “A tale of three constitutions: Ethnicity and politics in Fiji,” International Journal of Constitutional Law, Vol. 5, No. 4, 2007. pp. 639-669.

Gounder, Christine, “Fiji 2000: Journalist and the George Speight coup,” Pacific Journalism Review, Vol. 13, No. 1, 2007. pp. 125-141.

Hafez, Farid, Heinisch, Reinhard, “Breaking with Austrian Consociationalism: How the Rise of Rightwing Populism and Party Competition Have Changed Austria’s Islam Politics*,” Politics and Religion, Vol. 11, 2018. pp. 649-678.

Halperin-Kaddari, Ruth, Yadgar, Yaacov, “Between Universal Feminism and Particular Nationalism: politics, religion and gender (in)equality in Israel,” Third World Quarterly, Vol. 31, No. 6, 2010. pp. 905-920.

Haque, M Shamsul, “The Role of the State in Managing Ethnic Tensions in Malaysia,” American Behavioral Scientist, Vol. 47, No. 3, November, 2003. pp. 240-266.

Hart, Vivien, “Democratic Constitution Making,” United States Institute of Peace, Special Report 107, July, 2003.

Hazan, Reuven Y., “Religion and politics in Israel: The rise and fall of the

CEU eTD Collection consociational model,” Israel Affairs, Vol. 6, No. 2, 1999. pp. 109-137.

Head, Michael, “A Victory for Democracy? An Alternative Assessment of Republic of Fiji v Prasad, Melbourne Journal of International Law, Vol. 2, 2001.

Henderson, Karen, “Czechoslovakia; The failure of consensus politics and the break- up of the federation,” Regional and Federal Studies, Vol. 5, No. 2, 1995. pp. 111-133.

Horowitz, Donald L., “Explaining the Northern Ireland Agreement: The Source of an Unlikely Constitutional Consensus,” British Journal of Political Science, Vol. 32, 2002. pp. 193-220.

Horowitz, Donald L., “Three Dimensions of Ethnic Politics,” World Politics, Vol. 23, No. 2, January, 1971. pp. 232-244.

Ishak, Mohamed Mustafa, “Managing Ethnicity and Constructing the ‘Bangsa Malaysia’ (A United Malaysian Nation),” Malaysian Management Journal, Vol. 6, No. 1&2. pp. 99-115.

Iyer, Venkat, “Enforced Consociationalism and Deeply Divided Societies: Some Reflections on Recent Developments in Fiji,” International Journal of Law in Context, Vol. 3, 2007. pp. 127-153

Jakala, Michaelina, Kuzu, Durukan, Qvortrup, Matt, “Consociationalism and Power- Sharing in Europe,” Palgrave Macmillan, 2018. First Edition

Jenne, Erin K., “The Paradox of Ethnic Partition: Lessons from de facto Partition in Bosnia and Kosovo,” Regional and Federal Studies, Vol. 19, No. 2, 2009. pp. 273-289.

Jinadu, L. Adele, “Federalism, the Consociational State, and Ethnic Conflict in Nigeria.” Publius, Vol. 15, No. 2, 1985. pp. 71-100.

Klitgaard, Robert, Katz, Ruth, “Overcoming Ethnic Inequalities: Lessons from Malaysia,” Journal of Policy Analysis and Management, Vol. 2, No. 3, Spring, 1983. pp. 333- 349.

Kook, Rebecca, “Representation, minorities and electoral reform: the case of the Palestinian minority in Israel,” Ethnic and Racial Studies, Vol. 40, No. 12, 2017. pp. 2039- 2057.

Kuipers, Ank, “Suriname’s Bouterse on track for second presidential term,” Reuters, 26 May, 2015. https://www.reuters.com/article/uk-suriname-election/surinames-bouterse-on- track-for-second-presidential-term-idUKKBN0OB1JW20150526

Kyriakou, Nikolas, Skoutaris, Nikos, “The Birth of a Republic, But Not of a Nation: The Case of State-Building in Cyprus,” Nationalism and Ethnic Politics, Vol. 22, No. 4, 2016. pp. 457-477.

Lemarchand, René, “Consociationalism And Power Sharing in Africa: Rwanda,

CEU eTD Collection Burundi, And the Democratic Republic of The Congo,” African Affairs, 2006. pp. 1-20.

Lewin-Epstein, Noah, Cohen, Yinon, “Ethnic origin and identity in the Jewish population of Israel,” Journal of Ethnic and Migration Studies, 2018. pp. 1-20.

Lijphart, Arend, “Consociation and Federation: Conceptual and Empirical Links,” Canadian Journal of Political Science, Vol. 12, No. 3, September, 1979. pp. 499-515.

Lijphart, Arend, “Consociational Democracy,” World Politics, Vol. 21, No. 2, January, 1969. pp. 207-225.

Lijphart, Arend, “Power-Sharing in South Africa,” Institute of International Studies University of Southern California, Berkley, 1985.

Lijphart, Arend, “South African democracy: Majoritarian or consociational?” Democratization, Vol. 5, No. 4, 1998. pp. 144-150.

Lijphart, Arend, “The Puzzle of Indian Democracy: A Consociational Interpretation,” The American Political Science Review, Vol. 90, No. 2, June, 1996. pp. 258-268.

Lijphart, Arend, “Verzuiling, pacificatie en kentering in de Nederlandse politiek,” Haarlem, 1990. 8th print.

Lipshits, Hadar Moshe, Neubauer-Shani, Michal, “The state-religion issue in Israel: is the consociational model still alive?” Israel Affairs, 2019. pp. 1-17.

Macek-Macková, Emanuela “Challenges in conflict management in multi-ethnic states – the dissolution of Czechoslovakia and Serbia and Montenegro,” Nationalitiy Papers, Vol. 39, No. 4, 2011. pp. 615-633.

Magone, José M., “The Statecraft of Consensus Democracies in a Turbulent World: A Comparative Study of Austria, Belgium, Luxembourg, the Netherlands and Switzerland,” Taylor & Francis, December, 2016.

Marchand, Iris, “Dogla politics? Questioning ethnic consociationalism in Suriname’s national elections of 25 May 2010,” Ethnic and Racial Studies, Vol. 37, No. 2, 2014. pp. 342- 362.

Mattes, Robert B., “South Africa: Democracy Without the People?” Journal of Democracy, Vol. 13, No. 1, January, 2002. pp. 22-36.

McCulloch, Allison, Vandeginste, Steff, “Veto power and power-sharing: insights from Burundi (2000–2018),” Democratization, 2019. pp. 1-18.

Mill, John Stuart, “Considerations on Representative Government,” Henry Holt & Co., New York, 1873.

Mishali-Ram, Meirva, “When Ethnicity and Religion Meet: Kinship Ties and Cross- Border Dynamics in the Afghan- Pakistani Conflict Zone,” Nationalism and Ethnic Politics,

CEU eTD Collection Vol. 17, No. 3, 2011. pp. 257-275

Noel, Sid, “From Power-Sharing to Democracy: Post-Conflict Institutions in Ethnically Divided Societies,” McGill-Queen’s Press, 22 September, 2005.

NOS, “Het aantal moslims stijgt, maar met hoeveel?” Binnenland, politiek, 14 March, 2017. https://nos.nl/artikel/2163084-het-aantal-moslims-stijgt-maar-met-hoeveel.html

Oostindie, Gert, “Dependence and autonomy in sub-national island jurisdictions: the case of the Kingdom of the Netherlands,” The Round Table, Vol. 95, No. 386, 2006. pp. 609- 626.

Parool, “Nederlandse Antillen opgeheven,” 10 October, 2010. https://www.parool.nl/nieuws/nederlandse-antillen-opgeheven~bd052a24/

Phinney, Jean S., Horenczyk, Gabriel, Liebkind, Karmela, Vedder, Paul, “Ethnic Identity, Immigration, and Well-Being: An Interactional Perspective,” Journal of Social Issues, Vol. 57, No. 3, 2001. pp. 439-510.

Rabushka, Alvin, Shepsle, Kenneth A., “Politics in Plural Societies: A Theory of Democratic Instability,” Charles E. Merrill Publishing Company, Columbus, Ohio, 1972. First edition.

Raffoul, Alexandre W., “The Politics of Association: Power-Sharing and the Depoliticization of Ethnicity in Post-War Burundi,” Ethnopolitics, 2018. pp. 1-18.

Reddy, Movindri, “Challenging Democracy: Ethnicity in Postcolonial Fiji and Trinidad,” Nationalism and Ethnic Politics, Vol. 17, No. 2, 2011. pp. 182-202.

Reynolds, Andrew, “Building Democracy After Conflict: Constitutional Medicine,” Journal of Democracy, Vol. 16, No. 1, January, 2005. pp. 54-68.

Sani, Mohd Azizuddin Mohd, “The Emergence of New Politics in Malaysia From Consociational to Deliberative Democracy,” Taiwan Journal of Democracy, Vol. 5, No. 2, December, 2009. pp. 99-125.

Schmeets, Hans, “De religieuze kaart van Nederland, 2010-2015,” CBS, December, 2016.

Seshia, Shaila, “Divide and Rule in Indian Party Politics: The Rise of the Bharatiya Janata Party,” Asian Survey, Vol. 38, No. 11, November, 1998. pp. 1036-1050.

Singh, Chaitram, “Suriname and the Limits of Consociationalism,” Journal of Third World Studies, Vol. 31, No. 1, 2014. pp. 131-148.

Singh, H. “Ethnic Conflict in Malaysia Revisited,” Commonwealth and Comparative Politics, Vol. 39, No. 1, 2001. pp. 42-65.

Staniševski, Dragan M., Miller, Hugh T., “The Role of Government in Managing

CEU eTD Collection Intercultural Relations; Multicultural Discourse and the Politics of Culture Recognition in Macedonia,” Administration & Society, Vol. 41, No. 5, September, 2009. pp. 551-575.

Stroschein, Sherril, “Consociational Settlements and Reconstruction: Bosnia in Comparative Perspective (1995-Present),” The Annals of the American Academy of Political and Social Science, Vol. 656, November, 2014. pp. 97-115.

Taylor, Andrew, “Electoral systems and the promotion of ‘consociationalism’ in a multi-ethnic society. The Kosovo Assembly elections of November 2001,” Electoral Studies, Vol. 24, 2005. pp. 435-463.

Tonge, Jonathan, “Northern Ireland: Conflict and Change,” Routledge Taylor and Francis Group, London and New York, 2002. Second edition.

Toyana, Mfuneko, “South African land reform panel recommends seizures without pay in certain circumstances,” Reuters, 28 July, 2019. https://www.reuters.com/article/us- safrica-land/south-african-land-reform-panel-recommends-seizures-without-pay-in-certain- circumstances-idUSKCN1UN0F9

Traniello, Marisa, “Power-Sharing: Lessons from South Africa and Rwanda,” International Public Policy Review, Vol. 3, No. 2, March, 2008. pp. 28-43.

Tweede Kamer der Staten-Generaal, “Fracties,” Kamerleden en Commissies, accessed 5 May, 2019. https://www.tweedekamer.nl/kamerleden_en_commissies/fracties

Uvin, Peter, “Ethnicity and Power in Burundi and Rwanda: Different Paths to Mass Violence,” Comparative Politics, Vol. 31, No. 3, April, 1999. pp. 253-271.

van Aller, H. B., “Van kolonie tot koninkrijksdeel : de staatkundige geschiedenis van de Nederlandse Antillen en Aruba van 1634 tot 1994,” Wolters-Noordhoff, 1994.

van der Vyver, Johan D., “Constitutional Options for Post-Apartheid South Africa,” Emory Law Journal, Vol. 40, 1991. pp. 745-835.

Vandeginste, Stef, “Power-Sharing, Conflict and Transition in Burundi: Twenty Years of Trail and Error,” Africa Spectrum, Vol. 44, No. 3, 2009. pp. 63-86.

Vatter, Adrian, “Switzerland on the Road from a Consociational to a Centrifugal Democracy?” Swiss Political Science Review, Vol. 22, No. 1, 2016. pp. 59-74.

Veenendaal, Wouter, “Does Smallness Enhance Power-Sharing? Explaining Suriname’s Multiethnic Democracy,” Ethnopolitics, 2019. pp. 1-21.

Zembla international, “Minister Blok: Suriname is a failed state due to ethnic division,” Zembla, 18 Juli, 2018. https://zembla.bnnvara.nl/nieuws/minister-stef-blok- suriname-is-a-failed-state-due-to-ethnic-division

CEU eTD Collection